Our hypothesis is that lean women with polycystic ovary syndrome will have decreased peak bone mass compared to lean normally cycling control women and obese women. We propose to examine bone density by dual X-ray absorptiometry in a group of reproductive age women with polycystic ovary symdrome (PCOS) compared to age, weight, and ethnicity matched control cycling women. The study will consist of a menstrual history, a physical to determine height, weight and waist hip ratio, and a vaginal or abdominal ultrasound to determine ovarian morphology. An ovarian morphology consistent with polycystic ovaries is not required to make a diagnosis, but serves as a bioassay of prolonged exposure to elevated androgens. Blood will be drawn to determine testosterone, free testrosterone, estrone and estradiol. Lumbar spine bone mineral density and total body bone mineral density will be determined with a bone absorptiometer. Lumbar and femur bone density measurements will also be obtained. This same tests will be used to calculate total body lean mass and total body fat mass.